Llama-4-Scout-17B-16E-Instruct

Quantized version of Llama-4-Scout-17B-16E-Instruct, by Unsloth, optimized for instruction tasks and runs via Ollama.

Run

ollama run yasserrmd/Llama-4-Scout-17B-16E-Instruct

⚠️ Requires ~43 GiB system RAM even with Q2_K_XL quantization.

Notes

• Format: Q2_K_XL (lightweight GGUF quant)
• Good for Q&A, summaries, code, and chat
• Built with Unsloth for efficient fine-tuning

Example

> Explain RAG in simple terms.